any other best way of reading the file

**psbasha** · Jan 5 '08, 07:48 AM

Originally posted by bvdet

This link has some good introductory and intermediate information on regular expressions - LINK

I have been using Kodos for experimenting and testing regular expressions and mostly learned by practicing with and incorporating into my scripts when needed. I do not consider myself an expert on re. Trial and error may be the hard way, but that's the way I learned what I know about Python.

Code:

Re
pattnum = re.compile(r'''
                      -\d+\.\d+E\+\d+|          # engineering notation -+
                      \d+\.\d+E\+\d+|           # engineering notation ++
                      -\d+\.\d+E-\d+|           # engineering notation --
                      \d+\.\d+E-\d+|            # engineering notation +-
                      -\d+\.\d+|                # negative float format
                      \d+\.\d+|                 # positive float format
                      -\d+\.|                   # negative float format
                      \d+\.|                    # positive float format
                      -\.\d+|                   # negative float format
                      \.\d+|                    # positive float format
                      \d{1,8}                   # positive integer
                      ''', re.X


key_patt = re.compile(r'/([A-Za-z_-]+)/')
data_patt = re.compile(r'\d+\.\d+|\d+|\w+')

Hi BV,

Can you please elaborate the explanation for the above pattern ,with simple examples.What each pattern line stands for?. How are we deciding to go for this types of pattern.

Thanks
PSB

**bvdet** · Jan 5 '08, 03:30 PM

Originally posted by psbasha

Code:

Re
pattnum = re.compile(r'''
                      -\d+\.\d+E\+\d+|          # engineering notation -+
                      \d+\.\d+E\+\d+|           # engineering notation ++
                      -\d+\.\d+E-\d+|           # engineering notation --
                      \d+\.\d+E-\d+|            # engineering notation +-
                      -\d+\.\d+|                # negative float format
                      \d+\.\d+|                 # positive float format
                      -\d+\.|                   # negative float format
                      \d+\.|                    # positive float format
                      -\.\d+|                   # negative float format
                      \.\d+|                    # positive float format
                      \d{1,8}                   # positive integer
                      ''', re.X


key_patt = re.compile(r'/([A-Za-z_-]+)/')
data_patt = re.compile(r'\d+\.\d+|\d+|\w+')

Hi BV,

Can you please elaborate the explanation for the above pattern ,with simple examples.What each pattern line stands for?. How are we deciding to go for this types of pattern.

Thanks
PSB

Each line in pattnum matches a slightly different format of number as noted in the comments. The last line (''', re.X) contins the VERBOSE flag, which tells the compiler to ignore unecsaped whitespace and comments. The next to last line (\d{1,8}) greedily matches between 1 and eight digits at a time. That is what we fixed earlier to work with your formatted data.

key_patt matches words like this:
/ABC_abc-def/
The brackets '[......]' tell the compiler to match the set of characters enclosed. Since the slash characters are outside the brackets, they must enclose the word in a given string to match. That's how we matched your keywords.

data_patt matches a floating point number, integer or alphanumeric character. The character '|' tells the compiler to match the patttern to the left OR the pattern to the right in a given string.

**psbasha** · Jan 6 '08, 03:01 AM

Originally posted by bvdet

You will need to make a small change to regex pattern pattnum:[code=Python]
pattnum = re.compile(r'''
-\d+\.\d+E\+\d+| # engineering notation -+
\d+\.\d+E\+\d+| # engineering notation ++
-\d+\.\d+E-\d+| # engineering notation --
\d+\.\d+E-\d+| # engineering notation +-
-\d+\.\d+| # negative float format
\d+\.\d+| # positive float format
-\d+\.| # negative float format
\d+\.| # positive float format
-\.\d+| # negative float format
\.\d+| # positive float format
\d{1,8} # positive integer
''', re.X
)[/code]This will prevent the matching of more than 8 digits at a time. Further adjustments may be required.

Code:

SampleData
Line1*  1               1                1              2
*       .002952         .992547         .121827
$

Rect2   2        1      2       3       7       6
Rect    3        1      3       4       8       7
PRect2* 4               11              15              16
*       10              11              0.3
Rect2*   4               1               5               6
*       10              11              0.
Othr*   1               1               5               6
*       10              11              0.              0.
*       10              11              0.              1.0
Oth1*   1               1               5               6
*       10              11              0.              0.
*       10              11              0.              1.0
*       10              11              0.              1.0
*       10              11              0.              1.0
Rect*   5               1               5               6
*       10              11              0.
Rect    10000000	10000000200000007000000060000000
Rect    20000000	20000000300000008000000070000000
Rect    30000000	30000000400000009000000080000000
Tria3   40000000	400000005000000090000000
$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$  $
Tria    6        1      7       2       11
$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$  
Point   1               0.0     0.0     0.0
Point   2               1.0     0.0     0.0
Point   3               2.0     0.0     0.0
Point   4               3.0     0.0     0.0
Point   5               0.0     1.0     0.0
Point   6               1.0     1.0     0.0
Point   7               2.0     1.0     0.0
Point   8               4.0     1.0     0.0
Point*  9                               0.0             2.0
*       0.0
Point  *3280504         0               1.28286145E+03  1.28286145E+03
*       -2.01004501E+02
Point  *3280505         0               1.28286145-03  1.28286145+03
*       -2.01004501+02
Point   10000000	0.      0.      0.
Point   20000000	5.      0.      0.
Point   30000000	10.     0.      0.
Point   40000000	15.     0.      0.
Point   50000000	20.     0.      0.
Point   60000000	0.      5.      0.
Point   70000000	5.      5.      0.
Point   80000000	10.     5.      0.
Point   90000000	15.     5.      0.
$
END

if the keywords are defined as below
keywords = ['Point', 'Othr', 'Rect2', 'Rect','PRect', 'PLine', 'Line1', 'Tria'3,'Oth1']
The output we are getting is Incorrect.

Code:

Output
$Incorrect output
Line1
    [1, 1, 1, 1, 2, 0.0029520000000000002, 0.99254699999999996, 0.121827]
Tria3
    [3,40000000, 40000000, 50000000, 90000000]

Oth1
    [1, 1, 1, 5, 6, 10, 11, 0.0, 0.0, 10, 11, 0.0, 1.0, 10, 11, 0.0, 1.0, 10, 11, 0.0, 1.0]

Rect2
    [2, 2, 1, 2, 3, 7, 6]
    [2, 4, 1, 5, 6, 10, 11, 0.0]
$Correct output is 
Line1
    [1, 1, 1, 2, 0.0029520000000000002, 0.99254699999999996, 0.121827]
Tria3
    [40000000, 40000000, 50000000, 90000000]

Oth1
    [ 1, 1, 5, 6, 10, 11, 0.0, 0.0, 10, 11, 0.0, 1.0, 10, 11, 0.0, 1.0, 10, 11, 0.0, 1.0]

Rect2
    [ 2, 1, 2, 3, 7, 6]
    [ 4, 1, 5, 6, 10, 11, 0.0]

The source code is taking the 'Rect2' keyword number '2' also.sinmilarly for Line1,Tria3

**bvdet** · Jan 6 '08, 02:58 PM

Try making adjustments to pattnum and pattkey:[code=Python]
# last line in pattnum
# matches integers of length between 1 and 8 digits,
# if not preceded by an alpha character
# matching as many repetitions possible
............... .(?<![a-zA-Z])\d{1,8} # positive integer

# matches keywords listed in kargs
# may or may not have a trailing asterisk
# there must be a word boundary both ends
....pattkey = re.compile('|'. join([r'\b(%s)\*?\b' % item for item in kargs]))[/code]

**psbasha** · Jan 6 '08, 05:05 PM

Originally posted by bvdet

Try making adjustments to pattnum and pattkey:[code=Python]
# last line in pattnum
# matches integers of length between 1 and 8 digits,
# if not preceded by an alpha character
# matching as many repetitions possible
............... .(?<![a-zA-Z])\d{1,8} # positive integer

# matches keywords listed in kargs
# may or may not have a trailing asterisk
# there must be a word boundary both ends
....pattkey = re.compile('|'. join([r'\b(%s)\*?\b' % item for item in kargs]))[/code]

I try to add the pattern as suggested at the last

Code:

Pat
pattnum = re.compile(r'''
                      -\d+\.\d+E\+\d+|          # engineering notation -+
                      \d+\.\d+E\+\d+|           # engineering notation ++
                      -\d+\.\d+E-\d+|           # engineering notation --
                      \d+\.\d+E-\d+|            # engineering notation +-
                      -\d+\.\d+|                # negative float format
                      \d+\.\d+|                 # positive float format
                      -\d+\.|                   # negative float format
                      \d+\.|                    # positive float format
                      -\.\d+|                   # negative float format
                      \.\d+|                    # positive float format
                      \d{1,8}|
                      \(?<![a-zA-Z])\d{1,8}  # positive integer
                      ''', re.X
                     )

and the line

pattkey = re.compile('|'. join([r'\b(%s)\*?\b' % item for item in kargs]))

I hope there is a syntac error in the pattern

\(?<![a-zA-Z])\d{1,8} # positive integer.

I am getting the following error

Code:

Error
  File "C:\\Sample.py", line 12, in ?
    pattnum = re.compile(r'''
  File "C:\Python24\lib\sre.py", line 180, in compile
    return _compile(pattern, flags)
  File "C:\Python24\lib\sre.py", line 227, in _compile
    raise error, v # invalid expression
error: unbalanced parenthesis

**bvdet** · Jan 6 '08, 07:26 PM

You copied the pattern incorrectly. The characters '\(' are interpreted as the literal character '(' hence the unbalanced parentheses. The pattern I suggested should replace the last line in pattnum, not appended to the end of pattnum.

**psbasha** · Jan 6 '08, 11:55 PM

Originally posted by bvdet

You copied the pattern incorrectly. The characters '\(' are interpreted as the literal character '(' hence the unbalanced parentheses. The pattern I suggested should replace the last line in pattnum, not appended to the end of pattnum.

I tried replacing the last line as shown below.But still iam not getting the results as expected

Code:

Pat
-\.\d+|                   # negative float format
\.\d+|                    # positive float format                                            
\\(?<![a-zA-Z])\d{1,8}  # positive integer

Code:

Output
>>> Point
    [0.0, 0.0, 0.0]
    [1.0, 0.0, 0.0]
    [2.0, 0.0, 0.0]
    [3.0, 0.0, 0.0]
    [0.0, 1.0, 0.0]
    [1.0, 1.0, 0.0]
    [2.0, 1.0, 0.0]
    [4.0, 1.0, 0.0]
    [0.0, 2.0, 0.0]
    [1282.8614500000001, 1282.8614500000001, -201.004501]
    [0.0012828614500000001, 1282.8614500000001, -201.004501]
    [0.0, 0.0, 0.0]
    [5.0, 0.0, 0.0]
    [10.0, 0.0, 0.0]
    [15.0, 0.0, 0.0]
    [20.0, 0.0, 0.0]
    [0.0, 5.0, 0.0]
    [5.0, 5.0, 0.0]
    [10.0, 5.0, 0.0]
    [15.0, 5.0, 0.0]
PLine
Tria
    []
Line1
    [0.0029520000000000002, 0.99254699999999996, 0.121827]
PRect
Rect2
    []
Oth1
    [0.0, 0.0, 0.0, 1.0, 0.0, 1.0, 0.0, 1.0]
Rect
    []
    [0.0]
    []
    []
    []
Othr
    [0.0, 0.0, 0.0, 1.0]

I am getting the ouput without IDs for some cases and some list are empty.

Can you help me with the pattern ,how to replace and add to the last lines.

Thanks
PSB

**bvdet** · Jan 7 '08, 01:30 AM

[code=Python]pattnum = re.compile(r'''
-\d+\.\d+E\+\d+ # engineering notation -+
| # or
\d+\.\d+E\+\d+ # engineering notation ++
| # or
-\d+\.\d+E-\d+ # engineering notation --
| # or
\d+\.\d+E-\d+ # engineering notation +-
| # or
-\d+\.\d+ # negative float format
| # or
\d+\.\d+ # positive float format
| # or
-\d+\. # negative float format
| # or
\d+\. # positive float format
| # or
-\.\d+ # negative float format
| # or
\.\d+ # positive float format
| # or
(?<![a-zA-Z])\d{1,8} # positive integer
''', re.X
)[/code][code=Python]pattkey = re.compile('|'. join([r'\b(%s)\*?\b' % item for item in kargs]))[/code]Output:[code=Python]>>> Point
[86010101, 0.0, 0.0, 0.0]
[86010102, 1.0, 0.0, 0.0]
[86010103, 2.0, 0.0, 0.0]
[86010104, 3.0, 0.0, 0.0]
[86010106, 0.0, 1.0, 0.0]
[96010104, 1.0, 1.0, 0.0]
[86010110, 2.0, 1.0, 0.0]
[86010201, 4.0, 1.0, 0.0]
[86010115, 0.0, 2.0, 0.0]
[96010403, 0, 1282.8614500000 001, 1282.8614500000 001, -201.004501]
[3280505, 0, 0.0012828614500 000001, 1282.8614500000 001, -201.004501]
[10000000, 0.0, 0.0, 0.0]
[20000000, 5.0, 0.0, 0.0]
[30000000, 10.0, 0.0, 0.0]
[40000000, 15.0, 0.0, 0.0]
[50000000, 20.0, 0.0, 0.0]
[60000000, 0.0, 5.0, 0.0]
[70000000, 5.0, 5.0, 0.0]
[80000000, 10.0, 5.0, 0.0]
[90000000, 15.0, 5.0, 0.0]
PLine
[1, 6, 1.5, 9.375, 0.001, 0.001]
Tria
[40000000, 40000000, 50000000, 90000000]
[5, 1, 7, 2, 11]
Tria3
[60000000, 80000000, 90000000, 10000000]
PRect
[4, 11, 15, 16, 10, 11, 0.2999999999999 9999]
Line
[10001101, 1, 1, 2, 0.0029520000000 000002, 0.9925469999999 9996, 0.121827]
Oth1
[1, 1, 5, 6, 10, 11, 0.0, 0.0, 10, 11, 0.0, 1.0, 10, 11, 0.0, 1.0, 10, 11, 0.0, 1.0]
Rect1
[6, 2, 7, 8, 11, 12, 1.7612000000000 001]
Rect
[10001102, 1, 2, 3, 7, 6]
[10001103, 1, 3, 4, 8, 7]
[10001104, 1, 5, 6, 10, 11, 0.0]
[5, 1, 5, 6, 10, 11, 0.0]
[10000000, 10000000, 20000000, 70000000, 60000000]
[20000000, 20000000, 30000000, 80000000, 70000000]
[30000000, 30000000, 40000000, 90000000, 80000000]
Othr
[1, 1, 5, 6, 10, 11, 0.0, 0.0, 10, 11, 0.0, 1.0]
>>> [/code]I don't know where you got the leading backslashes. I made up some of the data for testing the patterns.

**psbasha** · Jan 7 '08, 06:22 AM

Thanks BV for your help.

I have to work more on RegEx.I have gone thru the link link you have shared with me.Still I have to work on that in detail.

How to verify whether our defined pattern is correct? .Whether we have to use KODOS to verify it or anyother links available

-PSB

**bvdet** · Jan 7 '08, 02:18 PM

Originally posted by psbasha

Thanks BV for your help.

I have to work more on RegEx.I have gone thru the link link you have shared with me.Still I have to work on that in detail.

How to verify whether our defined pattern is correct? .Whether we have to use KODOS to verify it or anyother links available

-PSB

You are welcome. The best way to test the patterns is thorough testing on real data. Remember that the keyword pattern requires the keyword to be a distinct word (there must be a space between the word and the first data field).

**psbasha** · Jan 9 '08, 10:54 PM

Originally posted by bvdet

[code=Python]pattnum = re.compile(r'''
-\d+\.\d+E\+\d+ # engineering notation -+
| # or
\d+\.\d+E\+\d+ # engineering notation ++
| # or
-\d+\.\d+E-\d+ # engineering notation --
| # or
\d+\.\d+E-\d+ # engineering notation +-
| # or
-\d+\.\d+ # negative float format
| # or
\d+\.\d+ # positive float format
| # or
-\d+\. # negative float format
| # or
\d+\. # positive float format
| # or
-\.\d+ # negative float format
| # or
\.\d+ # positive float format
| # or
(?<![a-zA-Z])\d{1,8} # positive integer
''', re.X
)[/code][code=Python]pattkey = re.compile('|'. join([r'\b(%s)\*?\b' % item for item in kargs]))[/code]Output:[code=Python]>>> Point
>>> [/code]I don't know where you got the leading backslashes. I made up some of the data for testing the patterns.

Code:

InputData
Point   5               0.0     1.0     0.0
Point   6               1.0     1.0     0.0
Point   7               2.0     1.0     0.0
Point   86010206        1471.51 -165.842139.
Point   86090104        1403.56 -148.237126.7
Point   86090129        1708.72 -722.4  232.274
Point   86120127        1664.09 -687.   225.852
$
END

BV,

Not able to fix the pattern for the above float format which has to be split at every '8' field.

The output is incorrect as shown
[86010206, 1471.51, -165.842139]
[86090104, 1403.5599999999 999, -148.23712599999 999, 0.6999999999999 9996]

It should be
[86010206, 1471.51, -165.842,139.0]
[86090104, 1403.56 -148.237, 126.7]

Need to get the X,Y,Z coordianets seperately.Your help is required here.

Thanks
PSB

**bvdet** · Jan 9 '08, 11:46 PM

Originally posted by psbasha

Code:

InputData
Point   5               0.0     1.0     0.0
Point   6               1.0     1.0     0.0
Point   7               2.0     1.0     0.0
Point   86010206        1471.51 -165.842139.
Point   86090104        1403.56 -148.237126.7
Point   86090129        1708.72 -722.4  232.274
Point   86120127        1664.09 -687.   225.852
$
END

BV,

Not able to fix the pattern for the above float format which has to be split at every '8' field.

The output is incorrect as shown
[86010206, 1471.51, -165.842139]
[86090104, 1403.5599999999 999, -148.23712599999 999, 0.6999999999999 9996]

It should be
[86010206, 1471.51, -165.842,139.0]
[86090104, 1403.56 -148.237, 126.7]

Need to get the X,Y,Z coordianets seperately.Your help is required here.

Thanks
PSB

The regex sequence pattnum depends on word boundaries except when matching integers. If your data consistently uses 8 character fields, the parsing would be much simpler.[code=Python]# Read data file with 8 character fields

import re

def convert_data(s) :
for func in (int, float):
try:
n = func(s)
return n
except:
pass
return s

def split_fields(s, fw=8):
outList = []
while s:
outList.append( s[:8])
s = s[8:]
return outList

def parse_data(fn, *kargs):
fileList = [item.strip() for item in open(fn).readli nes()\
if not item.startswith ('$')]

pattkey = re.compile('|'. join([r'\b(%s)\*?\b' % \
item for item in kargs]))

# create dictionary with keys from kargs
mDict = dict(zip(kargs, [[] for _ in kargs]))
for line in fileList:
m = pattkey.match(l ine)
if m:
lineList = [s.strip() for s in split_fields(li ne)]
mDict[m.group(0)].append([convert_data(it em) for item \
in lineList if item][1:])

return mDict

if __name__ == '__main__':
fn = r'H:\TEMP\temsy s\sample_points 100.txt'
dd = parse_data(fn, *['Point',])
for key in dd:
print key
for item in dd[key]:
print ' %s' % item

[/code]Output:[code=Python]>>> Point
[5, 0.0, 1.0, 0.0]
[6, 1.0, 1.0, 0.0]
[7, 2.0, 1.0, 0.0]
[86010206, 1471.51, -165.84200000000 001, 139.0]
[86090104, 1403.5599999999 999, -148.23699999999 999, 126.7]
[86090129, 1708.72, -722.39999999999 998, 232.274]
[86120127, 1664.0899999999 999, -687.0, 225.852]
>>> [/code]

**psbasha** · Jan 10 '08, 12:02 AM

>>> [/code][/QUOTE]
Sorry BV,

You have understood wrongly.My file data has 8 or 16 digits or combination of both filed formats.I have given a sample data that is not able to read correctly by the Pattern definition.
To support that we need to define the pattern for X,Y and Z coordinates properly for '8' and '16' digit.

Could you please help me in fixing this problem.

Thanks
PSB

**psbasha** · Jan 10 '08, 12:44 AM

Originally posted by psbasha

>>> [/code]

Sorry BV,

You have understood wrongly.My file data has 8 or 16 digits or combination of both filed formats.I have given a sample data that is not able to read correctly by the Pattern definition.
To support that we need to define the pattern for X,Y and Z coordinates properly for '8' and '16' digit.

Could you please help me in fixing this problem.

Thanks
PSB[/QUOTE]

Please find the Sample input data.

Code:

InputFileData
SampleData
Line1*  1               1                1              2
*       .002952         .992547         .121827
$
 
Rect2   2        1      2       3       7       6
Rect    3        1      3       4       8       7
PRect2* 4               11              15              16
*       10              11              0.3
Rect2*   4               1               5               6
*       10              11              0.
Othr*   1               1               5               6
*       10              11              0.              0.
*       10              11              0.              1.0
Oth1*   1               1               5               6
*       10              11              0.              0.
*       10              11              0.              1.0
*       10              11              0.              1.0
*       10              11              0.              1.0
Rect*   5               1               5               6
*       10              11              0.
Rect    10000000    10000000200000007000000060000000
Rect    20000000    20000000300000008000000070000000
Rect    30000000    30000000400000009000000080000000
Tria3   40000000    400000005000000090000000
$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$    $
Tria    6        1      7       2       11
$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$$    
Point   1               0.0     0.0     0.0
Point   2               1.0     0.0     0.0
Point   3               2.0     0.0     0.0
Point   4               3.0     0.0     0.0
Point   5               0.0     1.0     0.0
Point   6               1.0     1.0     0.0
Point   7               2.0     1.0     0.0
Point   8               4.0     1.0     0.0
Point*  9                               0.0             2.0
*       0.0
Point  *3280504         0               1.28286145E+03  1.28286145E+03
*       -2.01004501E+02
Point  *3280505         0               1.28286145-03  1.28286145+03
*       -2.01004501+02
Point   10000000        0.      0.      0.
Point   20000000        5.      0.      0.
Point   30000000        10.     0.      0.
Point   40000000        15.     0.      0.
Point   50000000        20.     0.      0.
Point   60000000        0.      5.      0.
Point   70000000        5.      5.      0.
Point   80000000        10.     5.      0.
Point   90000000        15.     5.      0.
Point   86010206        1471.51 -165.842139.
Point   86090104        1403.56 -148.237126.7
Point   86090129        1708.72 -722.4  232.274
Point   86120127        1664.09 -687.   225.852
$
END

**bvdet** · Jan 10 '08, 01:11 AM

Originally posted by psbasha

Sorry BV,

You have understood wrongly.My file data has 8 or 16 digits or combination of both filed formats.I have given a sample data that is not able to read correctly by the Pattern definition.
To support that we need to define the pattern for X,Y and Z coordinates properly for '8' and '16' digit.

Could you please help me in fixing this problem.

Thanks
PSB

The only thing I can think of is to correct the data on the fly. Before extracting all the numbers, split the string and test each list element to see if it has two or more decimal points. If True, add a word boundary (space character or other delimiter), then extract the data.[code=Python]
.......snip.... ....
'''
Before extracting all the numbers with pattnum, split the string and test
each list element to see if it has two or more decimal points. If True,
add a word boundary (space character or other delimiter), then extract
the data.
'''
def chk_str(s, fw=8, delim=' '):
sList = s.split()
for i, item in enumerate(sList ):
if item.count('.') > 1:
sList[i] = fix_fields(item , fw, delim)
return delim.join(sLis t)

def fix_fields(item , fw, delim):
offset = 0
for i in range(item.coun t('.')):
item = '%s%s%s' % (item[:8+offset], delim, item[8+offset:])
offset += 9
return item

def parseData(fn, *kargs):
fileList = [item.strip() for item in open(fn).readli nes()\
if not item.startswith ('$')]

pattkey = re.compile('|'. join([r'\b(%s)\*?\b' % item for item in kargs]))

# create dictionary with keys from kargs
masterDict = dict(zip(kargs, [[] for _ in kargs]))
inData = False
for line in fileList:

# test data string for missing word boundary between floating point numbers
# field width = 8
line = chk_str(line, 8)

# check for invalid data
if pattinvalid.sea rch(line):
for item in pattinvalid.fin dall(line):
line = line.replace(it em, item.replace('-', 'E-').replace('+', 'E+'))
.......snip.... ....[/code]

any other best way of reading the file

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment